Cerence has released an upgraded version of Cerence Drive, its flagship smart car virtual assistant platform. Cerence Drive provides car manufacturers with speech and voice recognition tech that can serve as a foundation for cloud-based and embedded conversational assistants.

The latest iteration of the platform makes improvements to the company’s natural language understanding (NLU) engine, which is built on one central technology stack and is compatible with 23 different languages. It also introduces a gender-neutral option for the Text-to-Speech (TTS) voice assistant. That same voice assistant will sound more human thanks to Cerence Drive’s updated deep learning tech, and can now be customized to change its speaking style and emotional delivery when conveying certain kinds of information.

The most noteworthy feature, however, may be Cerence Drive’s enhanced activation tech. The platform will listen for (and follow) specific questions or commands, negating the need for a button or a wake-up word to provide a more natural user experience.  

The new Cerence Drive will also do a better job of cancelling out background noise and identifying individual speakers throughout the entire vehicle.

“Cerence Drive is the driving force behind our promise of a truly moving in-car experience for our customers and their drivers, and our new innovations announced today are core to making that mission a reality,” said Cerence CEO Sanjay Dhawan.

Cerence’s technology has already appeared in a slew of smart car virtual assistants from major manufacturers like BMW and Daimler. The company is currently preparing to release its fiscal report for the fourth quarter, which will be the company’s first quarterly report since the company spun off from Nuance Communications in October.

Cerence recently formed a research partnership with Quebec’s Mila Artificial Intelligence Institute, and later highlighted the role of its three DRIVE Labs in its own research and development process.